Timeline for How to transform continuous data with extreme bimodal distribution
Current License: CC BY-SA 3.0
15 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Dec 19, 2014 at 10:07 | answer | added | Glen_b | timeline score: 5 | |
| Nov 17, 2014 at 10:39 | comment | added | Nick Cox | Incidentally, the fact the people don't (usually!) try to convert binary predictors to normal underlines that normality of predictors is not required. | |
| Nov 16, 2014 at 15:19 | comment | added | gung - Reinstate Monica | There appear to be 5 intermediate valued points, such that your variable is not perfectly binary. Do you know what the story is with those? | |
| Nov 14, 2014 at 21:55 | comment | added | user60721 | When would it be obvious to categorize it? This is truly the crux of my question. thanks in advance | |
| Nov 14, 2014 at 21:13 | comment | added | Nick Cox | I would leave it as is until and unless it becomes obvious that you should categorise it. (What you should do if you want to apply cluster analysis is a different question.) It does not matter how you plot it; this variable can't be made normal (or Gaussian, as I would prefer to call it). | |
| Nov 14, 2014 at 20:37 | history | edited | user60721 | CC BY-SA 3.0 |
deleted 21 characters in body
|
| Nov 14, 2014 at 20:34 | history | rollback | user60721 |
Rollback to Revision 2
|
|
| Nov 14, 2014 at 20:32 | history | edited | user60721 | CC BY-SA 3.0 |
added 81 characters in body
|
| Nov 14, 2014 at 20:30 | comment | added | user60721 | oh, okay. Should I forget about transforming and categorize it? I would like to perform cluster analysis on the data set. I have also included another distribution plot above based on stats.stackexchange.com/questions/25568/…. | |
| Nov 14, 2014 at 20:09 | comment | added | Nick Cox | Thanks for clarifying: this variable is not the response. (I guess @user777 like me thought that it was.) I would worry less about it then. Just proceed carefully and use lots of graphics to watch for side-effects. | |
| Nov 14, 2014 at 20:07 | history | edited | Nick Cox | CC BY-SA 3.0 |
deleted 7 characters in body
|
| Nov 14, 2014 at 19:26 | comment | added | user60721 | My outcome is dichotomous/binary with the following frequency 74% and 26% for "no" and "yes" respectively. I also have 36 potential predictor variables about 90% of these variable are categorical. | |
| Nov 14, 2014 at 18:24 | comment | added | Nick Cox | Two spikes will remain two spikes with any monotonic transformation. The good news is that regression does not require any marginal distribution to be normal. The bad news is that regression may not be a good method if the response variable has this kind of distribution. But why do you call this dichotomous? Dichotomous is not another word for bimodal. It means two, and only two, distinct values. | |
| Nov 14, 2014 at 18:16 | review | First posts | |||
| Nov 14, 2014 at 18:16 | |||||
| Nov 14, 2014 at 18:14 | history | asked | user60721 | CC BY-SA 3.0 |